The hypotheses to be studied for cancer of the extrahepatic bile duct include association with ulcerative colitis, cholecystectomy, benign disorders of the bile ducts, female hormone use, diet, and occupation. Since these cancers are rare, a case-control study is proposed. Cases will be inpatients in 11 major hospitals in eastern Massachusetts and Rhode Island. Probable cases will be interviewed but only histologically confirmed cases will be included in the analysis. 120 cases of the bile duct cancer are expected over the 4 and a half years of interviewing. Three control groups will be used: 1) neighborhood controls. For Massachusetts residents, controls will be selected from the town books and matched one-for-one to the cases on age, sex, and area of residences. For cases in Rhode Island the controls will be selected by random digit dialing. 2) Hospital controls. They will be inpatients in the same hospital at the time the case is interviewed, who have a diagnosis of benign neoplasms or other malignancies (excluding cancer of the digestive or respiratory systems or urinary bladder) and are residents of Massachusetts and Rhode Island. 3) In-law controls. Brothers-, sisters-, and cousins-in-law of the cases will be the third control group. Their inclusion is intended for the investigations of feasibility of such controls in a case-control study. Information will be collected from each subject on age, sex, race, ethnic origin, occupation, past medical history, body weight and height, parity, female hormone use, tobacco, alcohol and other beverage consumption, and diet. Standard analytic methods for matched and nonmatched case- control studies will be performed.